Read on Twitter
Hello world! I am excited to share my first ever publication entitled “BRCA1/BARD1 site-specific ubiquitylation of nucleosomal H2A is directed by BARD1,” now published online @NatureSMB . https://go.nature.com/3dfS5fg . Thread below:
#nucleosome #chromatin #ubiquitin
#nucleosome #chromatin #ubiquitin
Ubiquitin is the largest histone PTM with roles in DNA break repair pathways and transcriptional regulation. BRCA1/BARD1 is one of three RING-type E3 ligases that mono-ubiquitylate unique lysine sites on H2A in nucleosomes.
All three E3s (BRCA1/BARD1, Ring1b/Bmi1, and RNF168) bind to both the nucleosome substrate and the E2~Ubiquitin conjugate via their RING domains, presenting a unique opportunity to investigate mechanisms of site-specific mono-ubiquitylation.
Previous studies have shown that RNF168 and Ring1b/Bmi1 ubiquitylate lysine residues on opposite poles of the nucleosome disc surface by binding to unique surfaces of the histone core and directing the RING-bound E2 towards the lysine to be modified.
Links to these excellent studies:
Ring1b/Bmi1: https://go.nature.com/2N7oJoH
RNF168: https://go.nature.com/2Nq18j1
Ring1b/Bmi1: https://go.nature.com/2N7oJoH
RNF168: https://go.nature.com/2Nq18j1
However, the lysine targets of BRCA1/BARD1 (Lys125/127/129) are in a fully disordered region of the H2A C-terminal tail fewer than ten residues away from the target of Ring1b/Bmi1, calling to question how this ligase specifically targets these disordered lysine residues.
To address this, we solved a cryo-EM structure of the BRCA1/BARD1/E2/nucleosome complex. Although the BRCA1 RING binds to the canonical nucleosome acidic patch surface, the BARD1 RING forms a novel hydrophobic interface with the H2B/H4 cleft.
The BARD1-histone interface constrains it closer to the histone surface than its counterpart in the Ring1b/Bmi1 complex. This causes the E2-binding RING domain of BRCA1 to tilt upwards, elevating the E2 away from the histone surface — a molecular teeter-totter!
We propose that this increased distance between the E2 active site and H2A Lys118/119 on the histone surface is too far for efficient ubiquitin transfer, explaining why H2A Lys119 (the target of Ring1b/Bmi1) is not modified by BRCA1/BARD1.
Although we do not observe the H2A C-term tail in the structure, enzymatic assays indicate that the native lysine targets of BRCA1/BARD1 are optimally positioned to reach the E2 active site in the complex, and that lysine residues on specialized H2A isoforms may also be targeted.
Next, we use NMR to demonstrate that the H2A C-term tail retains flexibility in the BRCA1/BARD1/E2/nucleosome complex, but is more restricted in the analogous Ring1b/Bmi1 complex. We propose that his conformational restriction may prevent off-target modification by Ring1b/Bmi1.
Finally, we show that full-length BRCA1/BARD1 at ~300 kDa with >75% intrinsic disorder ubiquitylates nucleosomes faster and binds stronger than the RING heterodimer. This implies the presence of additional functional interactions that we are excited to reveal!
Our study completes the saga of the three known H2A-modifying RING E3s, and explains possible mechanisms of BARD1 RING mutations found in patients with cancer. It also highlights a critical function of the BARD1 RING, as BRCA1 tends to hog the spotlight!
Thank you to co-authors on Twitter (that I know of) @AniBurr @bluufive @kollmanlab @ChampChatt @KangJianming @AlexPravat @zhao_weixing, as well as those without! It was a wonderful team effort.
One more thing I forgot to mention, for the real chromatin aficionados: We observe H3 Lys79 methylation is inhibitory of BRCA1/BARD1-dependent nucleosome ubiquitylation, hinting at PTM crosstalk. This is due to its proximity to the BARD1-histone interface.
